Obesity is a major and growing health problem in the US. It affects more than 50 million Americans, causes significant morbidity (e.g. diabetes, hypertension), and contributes to increased mortality. New methods are thus much needed for delivering effective weight management serves to large numbers of people at a reasonable cost. The present research is designed to evaluate such weight management systems within the context of a managed care organization (MCO). A cohort of 1500 overweight members of an MCO will be identified and followed for 2 years. Five hundred (500) will be randomized to each of three experimental groups: 1) "usual care" control, 2) mail-based intervention comprised of an interactive program of weight counseling through the mail, and 3) phone-based intervention comprised of an interactive program of weight counseling by phone. Participation in available programs will be optional. The weight management systems will be evaluated in terms of the proportion of members who choose to enroll and mean weight change. Program costs and medical care utilization will also be examined. It is hypothesized that MCO members assigned to both the phone-based and mail- based groups will enroll in those weight control programs at higher rates than members assigned to the control group do to programs available as "usual care." It is also hypothesized that mean weight losses over 2 years among members who are offered mail-based and phone-based programs will be greater than in the control group. The proposed research is important because it focuses on long-term management of an important health problem in an applied setting (MCO) where such systems of long-term care might be realistically used. It also addresses important issues of cost and population outreach, and would provide a basis for determining whether such weight management systems are economically viable for primary health care providers.